While Bayesians do not like classical P-values and prefer measuring evidence in data through posterior probabilities of parameters or models, some problems like testing or exploration of goodness of fit of a single given model have led to the introduction of P-values. We confine ourselves to this particular context of goodness of fit in the brief discussion of Bayesian P-values. Most of this material is taken from Ghosh, Delampady and Samanta (2006) and Ghosh, Purkayastha and Samanta (2005).
Suppose that we have a single model M that specifies a density f(x | θ), θ ∈ Θ for the observable X and the Bayesian has a prior π(θ). The Bayesian wishes to examine how well the model M fits the data x obs on the basis of a statistic T(X) which measures the goodness of fit of data and model. Of course, T is also chosen by the Bayesian even though it is not part of the usual paradigm for Bayesian inference.
Let
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References and Further Reading
Bayarri MJ, Berger J (1998) Quantifying surprise in the data and model verification. In Bernardo JM et al. (eds) Bayesian statistics 6, Oxford University Press, Oxford, pp 53–82
Bayarri MJ, Berger JO (2000) P values for composite null models. J Am Stat Assoc 95:1127–1142, 1157–1170 (discussion)
Box GEP (1980) Sampling and Bayes inference in scientific modeling and robustness. J R Stat Soc (Series A) 143:383–430
Gelman A, Meng X, Stern H (1996) Posterior predictive assessment of model fitness via realized discrepancies. Stat Sinica 6:733–807 (with discussion)
Ghosh JK, Delampady M, Samanta T (2006) An introduction to Bayesian analysis: theory and methods. Springer, New York
Ghosh JK, Purkayastha S, Samanta T (2005) Role of P-values and other measures of evidence in Bayesian analysis. In: Dey DK, Rao CR (eds) Handbook of statistics, 25, Bayesian thinking: modeling and computation. pp 151–170
Guttman I (1967) The use of the concept of a future observation in goodness-of-fit problems. J R Stat Soc (Series B) 29:104–109
Meng XL (1994) Posterior predictive p-values. Ann Stat 22:1142–1160
Robins JM, van der Vaart A, Ventura V (2000) Asymptotic distribution of P Values for composite null models. J Am Stat Assoc 95:1143–1157, 1157–1170 (discussion)
Rubin DB (1984) Bayesianly justifiable and relevant frequency calculations for the applied statistician. Ann Stat 12:1151–1172
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this entry
Cite this entry
Ghosh, J.K., Delampady, M. (2011). Bayesian P-Values. In: Lovric, M. (eds) International Encyclopedia of Statistical Science. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-04898-2_136
Download citation
DOI: https://doi.org/10.1007/978-3-642-04898-2_136
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-04897-5
Online ISBN: 978-3-642-04898-2
eBook Packages: Mathematics and StatisticsReference Module Computer Science and Engineering